Dewpoint and Vector-Borne Disease Range Shifts

ISEF Category: Earth and Environmental Sciences

Subcategory: Other  ·  Difficulty: Advanced  ·  Setup: Home Setup  ·  Time: Full Year

The Hook

A few degrees of moisture in the air can change where mosquitoes and ticks survive. That means dewpoint, not just temperature, may help predict where disease risk moves next. If you can map that shift at the county scale, you turn climate data into a public health story. This kind of project sits right where Earth science meets real-world impact.

What Is It?

This project asks whether changing dewpoint tracks changes in the places where disease vectors, like mosquitoes and ticks, can spread. Dewpoint is the temperature at which air holds so much water vapor that condensation starts. Higher dewpoint usually means stickier, more humid air, and that can affect how long insects survive, how active they are, and where they can live.

Think of the climate as the neighborhood and the vectors as house guests with very picky needs. If the air gets too dry, some species struggle. If moisture rises, their habitat can expand. Your job is to test whether county-level dewpoint trends line up with reported West Nile, Lyme, or Aedes-vector patterns, then compare that link against other climate variables like temperature and precipitation.

Why This Is a Good Topic

This is a strong science fair topic because you can test a real climate hypothesis with free data and clear numbers. You do not need a wet lab, and you can still build an original model by choosing your own counties, time windows, variables, and validation method. It connects to public health, climate change, and disease ecology, so the work feels current and useful. You can also learn data cleaning, GIS mapping, statistics, and model checking, which are all valuable research skills.

Research Questions

• How does county-level dewpoint trend relate to reported West Nile disease incidence over time?
• What is the effect of average summer dewpoint on county-level Lyme disease reports after controlling for temperature?
• Does including dewpoint improve a climate-based model of Aedes-vector range shifts compared with temperature alone?
• To what extent do counties with faster dewpoint increases show larger changes in vector-borne disease reporting?
• Which climate variable, dewpoint, temperature, or precipitation, best predicts county-level mosquito-borne disease patterns?
• How does the relationship between dewpoint and disease reporting change across different U.S. regions?

Basic Materials

• Computer with internet access and enough storage for climate and county datasets.
• Spreadsheet software or Google Sheets for cleaning and sorting data.
• Python or R for analysis and plotting.
• Free GIS software such as QGIS for mapping county patterns.
• CDC ArboNET data for mosquito-borne disease reports.
• NOAA climate data or CMIP6 climate projections for dewpoint-related variables.
• U.S. Census county shapefiles for mapping counties.
• Notepad or document editor for tracking variable definitions and data sources.

Advanced Materials

• Computer with internet access and strong memory for large raster and county datasets.
• Python environment with geopandas, pandas, statsmodels, scipy, and matplotlib.
• R environment with sf, dplyr, ggplot2, and spatial modeling packages.
• QGIS or ArcGIS Pro for spatial joins and map production.
• CMIP6 downscaled climate projections from a trusted climate data portal.
• CDC ArboNET, CDC WONDER, or state surveillance data for validation.
• NOAA or USGS climate normals for comparison baselines.
• County adjacency or land-use layers for adding spatial context.
• High-resolution population data for rate normalization.
• Version control such as Git for tracking code changes.

Software & Tools

• Python: Cleans climate and disease datasets, runs regression models, and makes plots.
• R: Handles spatial analysis and statistical tests for county-level patterns.
• QGIS: Maps county data and helps you compare geographic trends.
• CDC ArboNET data tables: Provides public mosquito-borne disease surveillance data for U.S. counties.
• NOAA Climate Data Online: Supplies climate observations that you can compare with your projections.

Experiment Steps

1. Define the disease outcome you will study, then decide whether you will focus on West Nile, Lyme, Aedes-related range shifts, or a comparison across all three.
2. Choose the county scale, the time window, and the climate variables you will test so your dataset stays manageable.
3. Build one clean county table that links climate values, disease reports, and basic county features like region or population.
4. Plan a baseline model first, then decide how you will test whether dewpoint adds predictive power beyond temperature and precipitation.
5. Design map-based checks and holdout tests so you can see whether the pattern repeats in counties your model did not train on.
6. Decide how you will turn your findings into a fair comparison, such as correlation, regression, or classification metrics.

Common Pitfalls

• Using raw case counts instead of rates, which makes large counties look riskier even when their per-person burden is lower.
• Mixing annual climate averages with seasonal disease reports, which can hide the time window when vectors actually respond.
• Comparing county data without cleaning geographic mismatches, which can break the link between climate grids and county boundaries.
• Treating dewpoint as the only driver, which can blur the effect of temperature, precipitation, land use, and population density.
• Building a model with no validation split, which makes the fit look stronger than it really is.

What Makes This Competitive

A stronger version of this project does more than map two variables side by side. You would test whether dewpoint improves prediction after controlling for temperature, precipitation, and county size. You could also compare regions, seasons, or vector types to see where the relationship changes. Careful validation, clean spatial methods, and a clear explanation of uncertainty can make the project feel much more like real research.

Project Variations

• Focus only on West Nile risk and test whether summer dewpoint predicts county-level incidence better than summer temperature.
• Swap disease counts for vector range proxies, then model where Aedes-suitable counties may expand under CMIP6 projections.
• Compare humid coastal counties with drier inland counties to see whether dewpoint matters more in one climate zone than another.

Learn More

• CDC ArboNET: Search the CDC site for U.S. arboviral surveillance tables and county-level reports.
• NOAA Climate Data Online: Find observed dewpoint, temperature, and precipitation records for U.S. locations.
• NASA Earthdata: Search for climate and remote sensing datasets that can help add environmental context.
• CMIP6 documentation from ESGF or a university climate portal: Read model background and find projected climate variables.
• PubMed: Search review articles on dewpoint, humidity, vector ecology, and climate-sensitive disease transmission.
• MIT OpenCourseWare: Look for free courses in statistics, GIS, or environmental data analysis that support your modeling work.